MeshCommunication::gather() #192

benkirk opened this Issue Jan 25, 2014 · 0 comments

1 participant

libMesh - C++ Finite Element Library member

A nice optimization when a mesh is to be serialized but only to one processor (e.g. for I/O) would be MeshCommunication::gather(rank) instead of MeshCommunication::allgather().

Should be pretty straightforward.

@benkirk benkirk was assigned Jan 25, 2014
@benkirk benkirk closed this Jan 28, 2014
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment